Digital Commons Network^™

A Multimodal And Multilevel Ranking Scheme For Large-Scale Video Retrieval, Steven C. H. Hoi, Michael R. Lyu

Research Collection School Of Computing and Information Systems

A critical issue of large-scale multimedia retrieval is how to develop an effective framework for ranking the search results. This problem is particularly challenging for content-based video retrieval due to some issues such as short text queries, insufficient sample learning, fusion of multimodal contents, and large-scale learning with huge media data. In this paper, we propose a novel multimodal and multilevel (MMML) ranking framework to attack the challenging ranking problem of content-based video retrieval. We represent the video retrieval task by graphs and suggest a graph based semi-supervised ranking (SSR) scheme, which can learn with small samples effectively and integrate …

Semi-Supervised Svm Batch Mode Active Learning For Image Retrieval, Steven Hoi, Rong Jin, Jianke Zhu, Michael R. Lyu

Research Collection School Of Computing and Information Systems

Active learning has been shown as a key technique for improving content-based image retrieval (CBIR) performance. Among various methods, support vector machine (SVM) active learning is popular for its application to relevance feedback in CBIR. However, the regular SVM active learning has two main drawbacks when used for relevance feedback. First, SVM often suffers from learning with a small number of labeled examples, which is the case in relevance feedback. Second, SVM active learning usually does not take into account the redundancy among examples, and therefore could select multiple examples in relevance feedback that are similar (or even identical) to …

Semi-Supervised Distance Metric Learning For Collaborative Image Retrieval, Steven Hoi, Wei Liu, Shih-Fu Chang

Research Collection School Of Computing and Information Systems

Typical content-based image retrieval (CBIR) solutions with regular Euclidean metric usually cannot achieve satisfactory performance due to the semantic gap challenge. Hence, relevance feedback has been adopted as a promising approach to improve the search performance. In this paper, we propose a novel idea of learning with historical relevance feedback log data, and adopt a new paradigm called “Collaborative Image Retrieval” (CIR). To effectively explore the log data, we propose a novel semi-supervised distance metric learning technique, called “Laplacian Regularized Metric Learning” (LRML), for learning robust distance metrics for CIR. Different from previous methods, the proposed LRML method integrates both …

An Office Survival Guide, M. Thulasidas

Research Collection School Of Computing and Information Systems

In the unforgiving, dog-eat-dog corporate jungle, when you find yourself in a new corporate setting, you need to be sure of the welcome. More importantly, you need to prove yourself worthy of it.

Measuring Novelty And Redundancy With Multiple Modalities In Cross-Lingual Broadcast News, Xiao Wu, Alexander G. Hauptmann, Chong-Wah Ngo

Research Collection School Of Computing and Information Systems

News videos from different channels, languages are broadcast everyday, which provide abundant information for users. To effectively search, retrieve, browse and track news stories, news story similarity plays a critical role in assessing the novelty and redundancy among news stories. In this paper, we explore different measures of novelty and redundancy detection for cross-lingual news stories. A news story is represented by multimodal features which include a sequence of keyframes in the visual track, and a set of words and named entities extracted from speech transcript in the audio track. Vector space models and language models on individual features (text, …

Self-Organizing Neural Models Integrating Rules And Reinforcement Learning, Teck-Hou Teng, Zhong-Ming Tan, Ah-Hwee Tan

Research Collection School Of Computing and Information Systems

Traditional approaches to integrating knowledge into neural network are concerned mainly about supervised learning. This paper presents how a family of self-organizing neural models known as fusion architecture for learning, cognition and navigation (FALCON) can incorporate a priori knowledge and perform knowledge refinement and expansion through reinforcement learning. Symbolic rules are formulated based on pre-existing know-how and inserted into FALCON as a priori knowledge. The availability of knowledge enables FALCON to start performing earlier in the initial learning trials. Through a temporal-difference (TD) learning method, the inserted rules can be refined and expanded according to the evaluative feedback signals received …

Context Modeling With Evolutionary Fuzzy Cognitive Map In Interactive Storytelling, Yundong Cai, Chunyan Miao, Ah-Hwee Tan, Zhiqi Shen

Research Collection School Of Computing and Information Systems

To generate a believable and dynamic virtual world is a great challenge in interactive storytelling. In this paper, we propose a model, namely evolutionary fuzzy cognitive map (E-FCM), to model the dynamic causal relationships among different context variables. As an extension to conventional FCM, E-FCM models not only the fuzzy causal relationships among the variables, but also the probabilistic property of causal relationships, and asynchronous activity update of the concepts. With this model, the context variables evolve in a dynamic and uncertain manner with the according evolving time. As a result, the virtual world is presented more realistically and dynamically.

Multi-View Ear Recognition Based On B-Spline Pose Manifold Construction, Zhiyuan Zhang, Heng Liu

Research Collection School Of Computing and Information Systems

In this work, multi-view ear recognition problems are examined in detail. A new multi-view ear recognition approach based on B-Spline pose manifold construction in discriminative projection space which is formed by null kernel discriminant analysis (NKDA) feature extraction is presented. Many experiments and comparisons are provided to show the effectiveness of our multi-view ear recognition approach.

An Experimental Study On U-Commerce Adoption: Impact Of Personalization And Privacy Concerns, H. Sheng, Fiona Fui-Hoon Nah, Keng Siau

Research Collection School Of Computing and Information Systems

Ubiquitous commerce (u-commerce) represents “anytime, anywhere” commerce. U-commerce can provide a high level of personalization, which can bring significant benefits to customers. However, privacy is a major concern to customers and an obstacle to the adoption of u-commerce. This research examines how personalization and context can impact customers’ privacy concerns as well as intention to adopt u-commerce applications. As u-commerce is new and emerging, we used the scenario-based approach to operationalize personalization and context in an experimental study. The experimental results show that the effects of personalization on customers’ privacy concerns and adoption intention are situation dependent.

Appropriation Of Privacy Management Within Social Networking Sites, Catherine Dwyer

Dissertations

Social networking sites have emerged as one of the most widely used types of interactive systems, with memberships numbering in the hundreds of millions around the globe. By providing tools for their members to manage an ever-changing set of relationships, social networking sites push a constant expansion of social boundaries. These sites place less emphasis on tools that limit social boundaries to enable privacy.

The rapid expansion of online social boundaries has caused privacy shockwaves. Privacy offline is enabled by constraints of time and space. Online, powerful search engines and long term digital storage means private data have no expiration …

Perception Gaps And The Adoption Of Information Technology In The Clinical Healthcare Environment, Karen Hare

Dissertations

Implementation of information systems has lagged in many areas of clinical healthcare for a variety of reasons. Economics, data complexity and resistance are among the often quoted roadblocks. Research suggests that physicians play a major part in the adoption, use and diffusion of information technology (IT) in clinical settings. There are also other healthcare professionals, clinical and non-clinical, who play important roles in making decisions about the acquisition of information technology. In addition to these groups there are information technology professionals providing the services required within the healthcare field. Finally within this group are those IT professionals who have sufficient …

The Impact Of Cultural Differences In Temporal Perception On Global Software Development Teams, Richard William Egan

Dissertations

This dissertation investigated the impact of cultural differences in temporal perception on globally dispersed software development teams. Literature and anecdotal evidence suggest that these temporal differences affect individual communication quality, which in turn will affect individual satisfaction and trust within global teams. Additionally, the temporal dispersion of the team was expected to affect an individual's sense of temporal disruption which, in turn, was expected to affect individual satisfaction and trust. Differences in temporal perception were expected to moderate this impact on perceived temporal disruption. A Fortune 100 Company that carried out software testing in Ireland, the United States, China and …

Platform Competition In Digital Systems: Architectural Control And Value Migration, C. Jason Woodard

Research Collection School Of Computing and Information Systems

Digital systems give rise to complex layered architectures in which products at one layer serve as platforms for applications and services in adjacent layers. Platform owners face a difficult balancing act. On one hand, they need to make their platforms attractive to potential complementors by mitigating the threat of architectural lock-in. On the other hand, platform owners must be careful not to give away too much too soon, or risk being unable to recoup their own investments. This paper presents an agent-based model that explores this tension at both the firm and industry levels. Computational experiments show that boundedly rational …

Ranking Search Results For Translated Content, Brian Edwin Hawkins

Theses and Dissertations

Translation Memory (TM) is a valuable tool that helps human translators in doing their job. TM consists of a collection of previously translated texts, called translation units, that may prove useful in the translation of new text. The main problem faced by translators who wish to take advantage of TM is that, although search tools do exist, there is no standardized way of effectively ranking search results. This thesis proposes a method for ranking TM search results together with a novel approach to efficiently finding common substrings that is used in the ranking process.

Machine Code Verification Using The Bogor Framework, Joseph R. Edelman

Theses and Dissertations

Verification and validation of embedded systems software is tedious and time consuming. Software model checking uses a tool-based approach automating this process. In order to more accurately model software it is necessary to provide hardware support that enables the execution of software as it should run on native hardware. Hardware support often requires the creation of model checking tools specific to the instruction set architecture. The creation of software model checking tools is non-trivial. We present a strategy for using an "off-the-shelf" model checking tool, Bogor, to provide support for multiple instruction set architectures. Our strategy supports key hardware features …

Characterizing Effective Auction Mechanisms: Insights From The 2007 Tac Mechanism Design Competition, Jinzhong Niu, Kai Cai, Simon Parsons, Enrico Gerding, Peter Mcburney

Publications and Research

This paper analyzes the entrants to the 2007 TAC Market Design competition. It presents a classification of the entries to the competition, and uses this classification to compare these entries. The paper also attempts to relate market dynamics to the auction rules adopted by these entries and their adaptive strategies via a set of post-tournament experiments. Based on this analysis, the paper speculates about the design of effective auction mechanisms, both in the setting of this competition and in the more general case.

An Assistive Navigation Paradigm For Semi-Autonomous Wheelchairs Using Force-Feedback And Goal Prediction, John H.C. Staton

Computer Science and Engineering Theses - Archive

As computer technology advances and facilitates an increased amount of autonomous sensing and control, the interface between the human and autonomous computer systems becomes increasingly important. This applies in particular in the realm of service robotics where robots with increasing levels of autonomy are constantly interacting with human users and where it is paramount to have an efficient way to convey user intentions and commands to the robot system, integrate them into the robot's action and control, and to indicate robot control choices to the user in a "natural", intuitive way. One particular example where semi-autonomous operation of a vii …

Jcat: A Platform For The Tac Market Design Competition, Jinzhong Niu, Kai Cai, Simon Parsons, Enrico Gerding, Peter Mcburney, Thierry Moyaux, Steve Phelps, David Shield

Publications and Research

No abstract provided.

Energy-Efficient Connected K-Coverage, Duty-Cycling, And Geographic Forwarding In Wireless Sensor Networks, Habib M. Ammari

Computer Science and Engineering Dissertations - Archive

With the fast advances in inexpensive sensor technology and wireless communications, the design and development of large-scale wireless sensor networks has become cost-effective and viable enough to attract the attention of a wide range of civilian, natural, and military applications, such as health and environmental monitoring, seism monitoring, and battlefields surveillance. The main challenge in the design of wireless sensor networks is the limited battery power of the sensors and the difficulty of replacing and/or recharging these batteries due to the nature of the monitored field, such as hostile environments, and cost. Thus, it is necessary that the sensors be …

A Multi-Level Biomedical Ontology-Enabled Broker: Dynamic Service-Based Data Source Integration, Sheng-Chieh Jack Fu

Computer Science and Engineering Dissertations - Archive

Web services have recently become a new trend for gathering biomedical information. However, it is not easy to integrate and obtain a concise/complete query result among thousands of service operations. In this dissertation, we propose a multi-level ontology-enabled service broker system for dynamically integrating web services in the biomedical domain (BioServiceBroker). By introducing multi-level modeling concepts and intra/inter level relationships, our approach facilitates more accurate modeling of biomedical ontologies, which leads to a better understanding of the data stored in various biological data sources as well as the services provided by the data sources. In addition, we incorporate temporal concepts …

An Influenza Virus Molecular Infection Model And Discrete Event, Stochastic Simulation., Richard Burke Squires

Computer Science and Engineering Theses - Archive

Influenza virus is responsible for the greatest pandemic in human history causing 20 - 40 million deaths worldwide during the 1918 flu season. In 1997 fears of a future pandemic arose with the discovery of a new strain of H5N1 avian influenza. In an effort to better understand the dynamics of influenza infection a new model of influenza virus infection at the molecular level has been developed. Comprising nineteen stages and five molecular types (mRNA, cRNA, vRNA, vRNP and proteins) the FluSim model consists of the major contributing pathogen factors to influenza infection. A discrete event, stochastic simulation based upon …

Secure Data Aggregation In Wireless Sensor Networks, Wei Zhang

Computer Science and Engineering Dissertations - Archive

Recent advances in micro-electro-mechanical systems (MEMS) technology and wireless communications technologies have enabled the deployment of wireless sensor networks (WSNs) in a plethora of applications, ranging widely from military surveillance to civilian applications. To protect the networks from different kinds of attacks, security in wireless sensor networks plays a crucial role and has received increased attention especially in the applications deployed in hostile environments, such as battlefield monitoring and home security. While extensive efforts have been devoted toward securing conventional networks, the stringent resource constraints, such as energy, communication and computation capability, etc., have often prevented their direct adoptions. As …

Data Dissemination Protocols In Wireless Sensor Networks : Design, Modeling And Security, Pradip De

Computer Science and Engineering Dissertations - Archive

Wireless sensor and actuator networks have been one of the stepping stones towards realizing a pervasive computing infrastructure. However, in a post-deployment scenario, transferring critical updates and reconfigurations on a network-wide scale is a non-trivial proposition. Wireless techniques provide the only means of communication, and consequently, an in-depth study of the multihop broadcast based communication paradigm and the design of efficient and reliable data dissemination protocols for reprogramming a sensor network is of paramount importance. In this dissertation, we initially focus on the formal modeling and performance analysis of broadcast-based data dissemination protocols in wireless sensor networks. The data propagating …

Incorporation Of Evidences Into An Intelligent Computational Argumentation Network For A Web-Based Collaborative Engineering Design System, Xiaoqing Frank Liu, Ekta Khudkhudia, Ming-Chuan Leu

Computer Science Faculty Research & Creative Works

Conflicts among the stakeholders are unavoidable in the process of collaborative engineering design. Resolution of these conflicts is a challenging task. In our previous research, a web based intelligent collaborative system was developed which provides decision-making support, using computational argumentation techniques. Enhancements were done to this system to incorporate the priorities of the stakeholders and to detect arguments that self conflict. As an effort to make this system more effective and more objective in the process of decision making, we develop a method to assess the effect of evidences in the argumentation network, using Dempster-Shafer theory of evidence and fuzzy …

Scalable Detection Of Semantic Clones, Mark Gabel, Lingxiao Jiang, Zhendong Su

Research Collection School Of Computing and Information Systems

Several techniques have been developed for identifying similar code fragments in programs. These similar fragments, referred to as code clones, can be used to identify redundant code, locate bugs, or gain insight into program design. Existing scalable approaches to clone detection are limited to finding program fragments that are similar only in their contiguous syntax. Other, semantics-based approaches are more resilient to differences in syntax, such as reordered statements, related statements interleaved with other unrelated statements, or the use of semantically equivalent control structures. However, none of these techniques have scaled to real world code bases. These approaches capture semantic …

Verifying Completeness Of Relational Query Answers From Online Servers, Hwee Hwa Pang, Kian-Lee Tan

Research Collection School Of Computing and Information Systems

The number of successful attacks on the Internet shows that it is very difficult to guarantee the security of online servers over extended periods of time. A breached server that is not detected in time may return incorrect query answers to users. In this article, we introduce authentication schemes for users to verify that their query answers from an online server are complete (i.e., no qualifying tuples are omitted) and authentic (i.e., all the result values are legitimate). We introduce a scheme that supports range selection, projection as well as primary key-foreign key join queries on relational databases. We also …

Virtual Aesthetics And Ethical Communication: Towards Virtuous Reality Design, David Williams

All Theses

This thesis argues that ethics can and should be applied to Second Life avatar design and behavior. Second Life is a unique virtual reality due to its connection to the physical world primarily through financial devices. Users buy and sell virtual and physical goods over these networks; the avatar, it is argued, is the primary instrument for persuasion in these contexts. Avatars facilitate a virtual aesthetic that is primarily 'natural.' By creating aesthetic avatars, the developers of Second Life enable audiences to affectively associate with other 'residents.' Not only is the avatar designed for aesthetic appeal, but it enables users …

Caps: Concurrent Automatic Programming System, Ken Kennedy

All Dissertations

In the past few years, the focus in microprocessors has shifted from increasing speed to creating processors
that contain multiple cores. In order to effectively use the new processors, concurrent specifications
and applications must be developed. Additionally, there are many applications that require the specifications
to be provably correct. CAPS (Concurrent Automatic Programming System) is designed to aid the user in
the creation, execution, and formal verification of concurrent specifications.
The specification language of CAPS (CAPSL) is a very high-level language designed for concurrency
and automatic conversion to a colored Petri net (CP-net). For each statement of the language, there …

A Verification System For Timed Interval Calculus, Chunqing Chen, Jin Song Dong, Jun Sun

Research Collection School Of Computing and Information Systems

Timed Interval Calculus (TIC) is a highly expressive set-based notation for specifying and reasoning about embedded real-time systems. However, it lacks mechanical proving support, as its verification usually involves infinite time intervals and continuous dynamics. In this paper, we develop a system based on a generic theorem prover, Prototype Verification System (PVS), to assist formal verification of TIC at a high grade of automation. TIC semantics has been constructed by the PVS typed higher-order logic. Based on the encoding, we have checked all TIC reasoning rules and discovered subtle flaws. A translator has been implemented in Java to automatically transform …

A Dynamic Trust Management Scheme To Mitigate Malware Proliferation In P2p Networks, Xuhua Ding, Wei Yu, Ying Pan

Research Collection School Of Computing and Information Systems

The surge of peer-to-peer (P2P) networks consisting of thousands of of hosts makes them a breeding ground for malware proliferation. Although some existing studies have shown that malware proliferation can pose significant threats to P2P networks, defending against such an attack is largely an open problem. This paper aims to develop the countermeasure that can effectively mitigate the malware proliferation while preserving P2P networks' performance. To this end, we propose a dynamic trust management scheme based upon localized trust evaluation and alert propagation which prevents innocent peers from downloading files from infected peers. Our analysis and experimental results show that …